EXIF Image Viewer flaw

[Terry Pinnell]

I've just started using the handy EXIF Image Viewer, from Michael
Kowalski, following its recommendation in a thread here by kapok last
November.
http://home.pacbell.net/michal_k/
At first sight it has some excellent features, and I will at least use
it to supplement other programs like IrfanView.

However, it appears to have one major flaw IMO. The EXIF Date/Time it
displays is not the *original* date, but 'DateTime' - presumably when
the photo was downloaded. For me, and I'd have guessed the majority of
users, the most important EXIF field of all is the time and date at
which the photo was taken! Especially as so many image editors still
manage to destroy it at the slightest opportunity.

I'm unsure if it will reach him, but I've attempted an email to the
author asking him to offer DateTimeOriginal to be set in Options. If
anyone else has a reliable contact email, I'd appreciate it please.

BTW, this is where that 'Time Taken' field appears in
IrfanView>Info>EXIF

DateTime 2005:04:09 09:55:43
YCbCrPositioning Centered
ExifOffset 226
ExposureTime 1/60 seconds
FNumber 4.20
ExposureProgram Not defined
ExifVersion 0220
DateTimeOriginal 2005:04:08 12:09:09 <----

Does anyone know the formal definition of that first line 'DateTime'
please? Is it, as I suspect, the download date? Or perhaps date of
downlaod or of subsequent modification, whichever is later?

In that example above, I downloaded the image, taken by my son of me
and uploaded to his blog. It shows a date one day later than when it
was taken. As he apparently uses that DateTime to suffix his file
names, I'm keen to resolve this ambiguity. I can be 100% confident of
that, as it was a photo of me, visiting my son on a particular date.)

 

[David Gilbert]

BTW, this is where that 'Time Taken' field appears in
IrfanView>Info>EXIF

DateTime 2005:04:09 09:55:43
YCbCrPositioning Centered
ExifOffset 226
ExposureTime 1/60 seconds
FNumber 4.20
ExposureProgram Not defined
ExifVersion 0220
DateTimeOriginal 2005:04:08 12:09:09 <----

Does anyone know the formal definition of that first line 'DateTime'
please? Is it, as I suspect, the download date? Or perhaps date of
downlaod or of subsequent modification, whichever is later?

I do not know the formal definition but looking at some of my
images it is not time of downloading but time of modification.

 

[Terry Pinnell]

I do not know the formal definition but looking at some of my
images it is not time of downloading but time of modification.

Thanks. But that poses a puzzle. I copied the file above, and
confirmed that the renamed copy still contained those same two EXIF
date field values in IV. Then I cropped it, saved, and then viewed the
EXIF data again. I expected the 'DateTime' to change to today's date,
consistent with 'modified date'. But it remains at 2005:04:09
09:55:43.

As a different puzzle, I found an example in this same folder of
photos of DateTime being *older* than DateTimeOriginal! As that must
surely be physically impossible, I'm assuming it may be a summer time
issue?

Even more baffling is that many of these photos have *identical* EXIF
values for DateTime and DateTimeOriginal. Which doesn't seem possible!
And, confusing me further, if I repeat the renaming and editing test
on one of those, the EXIF values remain unaltered. So, DateTime cannot
mean 'time of modification'. In which case, what the heck *does* it
mean?

 

[David Gilbert]

Thanks. But that poses a puzzle. I copied the file above, and
confirmed that the renamed copy still contained those same two EXIF
date field values in IV. Then I cropped it, saved, and then viewed the
EXIF data again. I expected the 'DateTime' to change to today's date,
consistent with 'modified date'. But it remains at 2005:04:09
09:55:43.

As a different puzzle, I found an example in this same folder of
photos of DateTime being *older* than DateTimeOriginal! As that must
surely be physically impossible, I'm assuming it may be a summer time
issue?

Even more baffling is that many of these photos have *identical* EXIF
values for DateTime and DateTimeOriginal. Which doesn't seem possible!
And, confusing me further, if I repeat the renaming and editing test
on one of those, the EXIF values remain unaltered. So, DateTime cannot
mean 'time of modification'. In which case, what the heck *does* it
mean?

Don't know whether this helps
From the EXIF2.2 spec. see http://www.exif.org/specifications.html

DateTime
The date and time of image creation. ** In this standard it is the
date and time the file was changed. ** The format is
"YYYY:MMD HH:MM:SS" with time shown in 24-hour format, and the date
and time separated by one blank character [20.H].

DateTimeOriginal
The date and time when the original image data was generated. For a
DSC the date and time the picture was taken are recorded. The format
is "YYYY:MMD HH:MM:SS" with time shown in 24-hour format, and the
date and time separated by one blank character [20.H].

DateTimeDigitized
The date and time when the image was stored as digital data. If, for
example, an image was captured by DSC and at the same time the file
was recorded, then the DateTimeOriginal and DateTimeDigitized will
have the same contents. The format is "YYYY:MMD HH:MM:SS" with time
shown in 24-hour format, and the date and time separated by one blank
character [20.H].

 

[Terry Pinnell]

Thanks. But that poses a puzzle. I copied the file above, and
confirmed that the renamed copy still contained those same two EXIF
date field values in IV. Then I cropped it, saved, and then viewed the
EXIF data again. I expected the 'DateTime' to change to today's date,
consistent with 'modified date'. But it remains at 2005:04:09
09:55:43.

As a different puzzle, I found an example in this same folder of
photos of DateTime being *older* than DateTimeOriginal! As that must
surely be physically impossible, I'm assuming it may be a summer time
issue?

Even more baffling is that many of these photos have *identical* EXIF
values for DateTime and DateTimeOriginal. Which doesn't seem possible!
And, confusing me further, if I repeat the renaming and editing test
on one of those, the EXIF values remain unaltered. So, DateTime cannot
mean 'time of modification'. In which case, what the heck *does* it
mean?

Don't know whether this helps
From the EXIF2.2 spec. see http://www.exif.org/specifications.html

DateTime
The date and time of image creation. ** In this standard it is the
date and time the file was changed. ** The format is
"YYYY:MMD HH:MM:SS" with time shown in 24-hour format, and the date
and time separated by one blank character [20.H].

DateTimeOriginal
The date and time when the original image data was generated. For a
DSC the date and time the picture was taken are recorded. The format
is "YYYY:MMD HH:MM:SS" with time shown in 24-hour format, and the
date and time separated by one blank character [20.H].

DateTimeDigitized
The date and time when the image was stored as digital data. If, for
example, an image was captured by DSC and at the same time the file
was recorded, then the DateTimeOriginal and DateTimeDigitized will
have the same contents. The format is "YYYY:MMD HH:MM:SS" with time
shown in 24-hour format, and the date and time separated by one blank
character [20.H].

Thanks, I had seen those, but they don't really explain the anomalies
I described. Also, what exactly is "The date and time of image
creation"? For DateTimeOriginal, "The date and time when the original
image data was generated" is explicitly defined (for a DSC) as the
time of taking the photo, as I'd assumed. So does that confirm (in the
familiar context) that 'DateTime' refers to downloading to the PC?

And do these definitions therefore mean that any change to any of
those 3 EXIF date/time fields due to any later *modification* of the
image (editing of any kind in an image editor) does *not* alter their
values? (No mention is made of 'modification'.)